Need some measurment advise

[mintos]

Trying to get into measuring speakers and need some general tips. I generally understand that measuring outdoors (away from boundaries) is a good idea. Here are some other questions

1. What distances should measurements be taken at?

2. Do I have to worry about measuring at a certain voltage or spl? Or just set the vol knob to a certain point and take all the measurements there is ok?

3. Do I point the mic to tweeter lvl or just any where on the speaker where my ears will be?

4. Does the quality of the amp used to drive the speakers matter? (Assuming any decent amp running at < 50% max capacity)

Onmimic Specific questions

1. I don't use the cd but use the play through sound card option for spl. Is that ok?

2. How do I effectively measure lower frequencies? I know it has something to do with gating.

Thanks

[Sydney]

1. What distances should measurements be taken at?

2. Do I have to worry about measuring at a certain voltage or spl?

the De facto standard practice is to feed 1 watt in and measure @ 1 meter.
For 8 ohm load that is 2.83V in, recalculate/adjust for other loads and other distances. There are techniques for close in measurement and response "stitching".

3. Do I point the mic to tweeter lvl or just any where on the speaker where my ears will be?

Depending on the design 1M may be too close to get a coherent signature of the entire speaker.

2. How do I effectively measure lower frequencies? I know it has something to do with gating.

If you eliminate the reflections ( that contaminate the signal ) then gating is not necessary.

[ryanbouma]

1. What distances should measurements be taken at?

1m is common. For a recent build, I required 2m. It depends on the size of the speaker. To measure a driver alone, 1m is usually always fine.

2. Do I have to worry about measuring at a certain voltage or spl? Or just set the vol knob to a certain point and take all the measurements there is ok?

2.83V will give you the commonly used SPL/2.83V/m. But any volume that is not to low as to let ambient noise comsume the signal is fine, generally.

3. Do I point the mic to tweeter lvl or just any where on the speaker where my ears will be?

Depends on your design. Generally yes. Point it at the listening axis.

4. Does the quality of the amp used to drive the speakers matter? (Assuming any decent amp running at < 50% max capacity)

Best to use something you know is linear, or the amp you plan to use to listen. Most amps are pretty linear. Make sure it doesn't clip. At 2.83V, any amp should be fine.

1. I don't use the cd but use the play through sound card option for spl. Is that ok?

Can't see why not.

2. How do I effectively measure lower frequencies? I know it has something to do with gating.

Thanks

That's not an omnimic question. But it's a good question. The further from a reflection you are (on a 10' ladder outside) the lower you can set you gate. You can probably even read the raw measurement and it'll be nearly anechoic. That's been my experience with a speaker 4m above ground. Indoors, you need to set the gate carefully just before the first reflection on the impulse response.

If you want to measure frequencies below your gate, you'll have to either get further from reflections (that gets totally impractacle), Ground plane, or nearfield.

Good luck.

[sam]

http://www.parts-express.com/pe/show...number=500-030

[jonpike]

+1 on the above!

Dr. D'Appolito's book is a very good reference on all the ins and outs of measuring, different techniques and why to use them, etc. It should be in your library if you're serious about measuring.

[ryanbouma]

I've got that book and I've hardly even cracked it open. Thanks for the reminder Sam, I have a 3 hour flight in a week. I'll do some reading

[ReissM]

Trying to get into measuring speakers and need some general tips. .......

2. How do I effectively measure lower frequencies? I know it has something to do with gating.

Most of your questions have already been answered above...

They also mentioned "near-field" testing. Let me give this a try...

Usually, your far-field tests (like 1 meter or 2 meters) are only valid down to approximately the reciprocal of the gated time. For example: a gated time of 5mS will provide good data down to about 200 Hz. So any measurement frequencies below 200 Hz are not valid. That's where near-field testing can help.

Place the microphone 1/4 inch from the woofer's cone and run a test. Open up the gate time to 250mS. (OmniMic has a check box to achieve this) Theoretically, any reflections from walls and ceiling will be so much lower in amplitude (compared to the direct sound going from woofer cone to microphone) that they will be inconsequential.

Next you splice the near-field measurement with the far-field measurement and voila! Full frequency measurements without an expensive anechoic chamber.

Look up ground plane testing also.

[rpb]

Trying to get into measuring speakers and need some general tips. I generally understand that measuring outdoors (away from boundaries) is a good idea. Here are some other questions

1. What distances should measurements be taken at?

2. Do I have to worry about measuring at a certain voltage or spl? Or just set the vol knob to a certain point and take all the measurements there is ok?

3. Do I point the mic to tweeter lvl or just any where on the speaker where my ears will be?

4. Does the quality of the amp used to drive the speakers matter? (Assuming any decent amp running at < 50% max capacity)

Onmimic Specific questions

1. I don't use the cd but use the play through sound card option for spl. Is that ok?

2. How do I effectively measure lower frequencies? I know it has something to do with gating.

Thanks

Some of the answers depend on why you are measuring. For x-over work, you really don't need to measure low frequencies. (Well, hardly ever.)

The exact spl of the sweeps doesn't matter. For 2-ways, measuring from about 3' works well. I hold my mic by hand, so I'm all over the place with mic location. It really doesn't matter all that much in the initial x-over stages.

For a 3-way, a little ingenuity will let you measure pretty much the same as a 2-way. A few tricks will get around some problems with a tall spread of drivers.

Pointing the mic at the tweeter, or at a point between the tweeter and mid works fine. By the time you are done, you may do both, and also some further above or below the tweeter axis.

For a 3-way, work the woofer to mid x-over independant of the tweeter. Point the mic at the centerline between the woofer and mid. When you get it close, measure the tweater and mid x-over by pointing at or near the tweeter. Disconect the woofer if you want. After the tweeter x-over is roughed out, you can measure all three at once.

[ReissM]

The exact spl of the sweeps doesn't matter. For 2-ways, measuring from about 3' works well. I hold my mic by hand, so I'm all over the place with mic location. It really doesn't matter all that much in the initial x-over stages.

Hmmmmm... well, be careful. I agree that the exact SPL doesn't matter, as long as you are consistent and measure all drivers without adjusting the volume control.

However, in order to properly use Passive Crossover Designer by Jeff Bagby (many of us here on TTF use PCD. It's great!) you will need to find the acoustic offset. I stumbled on this step earlier this year. The distance during multiple measurements MUST be exactly the same. The phase relationships inherent in the measurements count on this fact.

If you're holding the mic with your hands... you're already corrupting your measurements' phase relationships and PCD will not sum exactly as expected. I'd suggest an inexpensive mic stand.

Not trying to call you out... I'm just sharing some of my own mistakes in this regard. When done carefully, the predictive capability of PCD will astound you. Check the forum for Jeff's suggestions on how to setup/initialize PCD with acoustic offset details.

[rpb]

Hmmmmm... well, be careful. I agree that the exact SPL doesn't matter, as long as you are consistent and measure all drivers without adjusting the volume control.

However, in order to properly use Passive Crossover Designer by Jeff Bagby (many of us here on TTF use PCD. It's great!) you will need to find the acoustic offset. I stumbled on this step earlier this year. The distance during multiple measurements MUST be exactly the same. The phase relationships inherent in the measurements count on this fact.

If you're holding the mic with your hands... you're already corrupting your measurements' phase relationships and PCD will not sum exactly as expected. I'd suggest an inexpensive mic stand.

Not trying to call you out... I'm just sharing some of my own mistakes in this regard. When done carefully, the predictive capability of PCD will astound you. Check the forum for Jeff's suggestions on how to setup/initialize PCD with acoustic offset details.

You are correct in regards to measuring for the purpose of using software for x-over simulation!

Part of my reply was to point out some differences depending on why you might be measuring. I use trial and error rather than PCD, so these things don't matter with trial and error. With trial and error, you don't need a computer program to tell you if the phase or offsets are correct. You measure the drivers together, and see if they sum the way you want them too.

[bbcmp1979]

rpb is right on the trial and error process if you have a lot of xover parts sitting around that you can swap with, and of course highly experienced too. Without parts sitting around, I would highly suggest PCD as a starting point for xover design and tweak from there.